This application addresses broad challenge area (05): Comparative Effectiveness Research and specific challenge topic 05-AG-101*: Data Infrastructure for Post-Marketing Comparative Effectiveness Studies. Data infrastructure is badly needed to conduct post-approval comparative effectiveness studies of drugs and other therapeutic agents. The investigators propose to leverage their previous work and established data resources by extending their current 1999-2005 database of five large Medicaid programs (California, Florida, New York, Ohio, and Pennsylvania) through 2007 (a total ever-enrolled population of 60 million beneficiaries), and obtaining and linking data from Medicare Parts A, B, and D on dual enrollees in those states. This will allow the study of Medicaid beneficiaries, including those co-enrolled in Medicare, the disabled, and nursing home residents. These newly linked data can serve as a national resource for comparative effectiveness research via research collaborations with the National Institutes of Health, the Food and Drug Administration, and three major national research networks: the National Institutes of Health-sponsored Clinical and Translational Science Award Consortium, the Agency for Healthcare Quality and Research (AHRQ)- funded Centers for Education and Research on Therapeutics (CERTs), and the AHRQ-funded Developing Evidence to Inform Decisions about Effectiveness (DEcIDE) Network. Further, to demonstrate the utility of this newly extended database, the investigators propose to perform drug utilization and comparative effectiveness research using these newly linked data. Specifically, the investigators will characterize overall medication use in Medicaid enrollees by identifying the most frequently dispensed therapeutic classes, both by prescription volume and by cost, for nursing home residents and nursing home non-residents, in Medicaid-only and Medicaid-Medicare beneficiaries. Then, to demonstrate the utility of this novel data resource for comparative effectiveness research, they will compare the effectiveness of different strategies for initial supplementation and monitoring of potassium in patients newly starting loop diuretics and those newly starting thiazide diuretics with regard to preventing 1) all-cause death and 2) sudden cardiac death / ventricular arrhythmia. PUBLIC HEALTH RELEVANCE: This study will result in an important, permanent, national data resource for conducting post-marketing comparative effectiveness studies. In addition, it will provide important drug utilization data for typically underrepresented populations, and address the comparative effectiveness of strategies for monitoring and supplementing serum potassium in patients receiving diuretics. By compiling a very large, longitudinal, and current data resource of Medicaid and Medicare health insurance claims (including Part D), this project will create an important, permanent, national data infrastructure for conducting post-marketing comparative effectiveness studies, thereby providing clinicians and patients with an evidence base for selecting among alternative therapies or approaches. To demonstrate the utility of this infrastructure, this project will examine drug utilization in Medicaid enrollees, including those also enrolled in Medicare, and evaluate differing strategies for supplementation and monitoring of potassium among diuretic users in preventing 1) all-cause death and 2) sudden cardiac death and ventricular arrhythmia, which are both serious outcomes of importance to patients, clinicians, and policy makers.